Post on 14-Jul-2020
1
Operational Amplifiers (Op Amps) E80 – 2/5/13 - Professor Katherine Candler
Notes courtesy of Professor Sarah Harris
Agenda • Ideal Op Amps • Real Op Amps
Ideal Op Amps Operational amplifiers (op amps) amplify an input signal and produce an output signal. The symbol for an op amp is shown below. Figure 1(a) shows the symbol with the power supplies, +
ccV and −ccV (also sometimes called +V and −V ) drawn explicitly. The power
supplies are typically V15± . Figure 1(b) shows a simplified symbol where the power supplies are implicit. The power supplies must always be connected to operate the circuit. +v and −v indicate the voltages of the two input ports. Note: some references refer to these two input voltages as +In and −In .
)( ++ VVcc
)( −− VVcc
outV
−v
+voutV
(a) (b)
−v
+v
Figure 1. Op amp symbols: (a) with explicit power supplies, (b) simplified Characteristics Characteristics that are useful in analyzing an op amp circuit are:
1. Current into +v and −v is 0. 2. Voltage at the two inputs is equal: +− = vv when circuit is configured in negative
feedback (i.e., the output is connected to the negative input terminal) 3. Input impedance looking into the input terminals is infinite: ∞=inR 4. Output impedance is small: 0≈outR
Let’s use these characteristics, particularly the first two, in analyzing the following example circuits.
2
Example 1: Inverter Consider the circuit in Figure 2. Because no current flows into either input of the op amp (characteristic 1), 21 ii = .
outvinv
1R
2R
1i
2i
−v
+v
Figure 2. Inverter Writing the currents in terms of voltages, we get:
21
21
Rvv
Rvvii
outin −=
−
=−−
Since the voltages at +v and −v are the same (characteristic 2) and +v = 0 V, −v = 0 V. Substituting into the previous equation, we get:
21 Rv
Rv outin −
=
1
2
RR
vv
in
out −=
So, for example, if 12 RR = , outv has the same magnitude but opposite phase of inv . If
12 RR > , the output is both amplified and inverted. Note that the voltage of the amplified output cannot exceed the voltage of the external supplies, V15± . We can also analyze the circuit in the frequency domain. Suppose 1R and 2R were replaced by generic impedances 1Z and 2Z as shown in Figure 3.
3
outVinV
1I
2I
−v
+v
1Z
2Z
Figure 3. Circuit with impedances
I1 = I2
Vin − v−
Z1=
v− −Vout
Z2
As before, since the voltages at +v and −v are the same (characteristic 2) and +v = 0 V,
−v = 0 V. Substituting into the previous equation, we get:
21 RV
RV outin −
=
1
2
ZZ
VV
in
out −=
Example 2: Voltage follower / buffer Consider the circuit in Figure 4. Because the voltages at +v and −v are the same (characteristic 2) and invv =+ , inout vvv == − .
outvinv
−v
+v
Figure 4. Voltage follower / buffer
Thus, 1=in
out
vv . The output voltage follows the input voltage. This device, where the
output is equal to the input, is also called a buffer. A buffer is useful for isolating the output from a noisy input, or vice versa. It also can convert a high output impedance circuit into a low output impedance circuit (characteristic 4).
4
Example 3: Circuit comparison with and without buffer Consider the circuits in Figure 5. Compare the output voltages of the two circuits measured by an oscilloscope with a 1 MΩ input impedance.
Ω= MR 101
Ω= MR 402
)2
sin(
5)(
tt
v sπ
=
Ω= 50sRSignal Generator
)(tvout Ω= MRinstr 1
Oscilloscope
Ω= MR 101
Ω= MR 402
)2
sin(
5)(
tt
v sπ
=
Ω= 50sRSignal Generator
)(tvoutΩ= MRinstr 1
Oscilloscope
1
2
(a) (b)
Figure 5. Comparison of circuit with and without buffer As we’ve analyzed before, the measured output voltage for the circuit in Figure 5(a) is:
vout =
R2 || Rinstr
(R2 || Rinstr )+ R1 + Rsvs ≈
Rinstr
Rinstr + R1vs =
111
vs
For the circuit in Figure 5(b), the impedance seen looking into node 1 is infinite (characteristic 4), so the voltage measured at node 1 is:
v1 =
R2R2 + R1 + Rs
vs ≈R2
R2 + R1vs =
45
vs
Since the output impedance of node 2 is approximately 0Ω (characteristic 3), all of the voltage delivered to node 2 is dropped across the 1 M Ω internal impedance of the oscilloscope, and the measured output voltage is:
sout vvv54
1 ==
Thus, inserting the buffer eliminated instrument loading effects.
5
Example 4: Differentiator Consider the circuit in Figure 6. We can analyze the circuit in the frequency domain using impedances and the previous derived results.
outVinV
R
Cjω1
1I
2I
−v
+v
Figure 6. Differentiator
RCjCj
RZZ
VV
RZCj
Z
in
out ωω
ω
−=−=−=
=
=
1
1
1
2
2
1
The output is the inverted derivative of the input. Example 5: Integrator Consider the circuit in Figure 7. We can analyze the circuit using impedances.
outVinV
R
Cjω1
1I
2I
−v
+v
Figure 7. Integrator
6
RCjRCj
ZZ
VV
CjZ
RZ
in
out
ωω
ω
11
1
1
2
2
1
−=−=−=
=
=
The output is the inverted integral of the input. Example 6: Differential amplifier Consider the circuit in Figure 8.
outV
2R
2I
1V1R
4R
2V3R
−v
+v
1I
Figure 8. Differential amplifier
21
1
21
RVv
RvVII
out−=
−
=−− 2
43
4 )( VRR
Rv+
=+
Substituting and after some algebra, we get:
])[(,
)())((
121
24231
11
22
1
21
43
4
VVRRVRRandRRwhen
VRRV
RRR
RRRV
out
out
−===
−+
+=
A differential amplifier amplifies the difference between two voltages. If all the resistance values are equal, the output is just the difference between the two input voltages. A differential amplifier is particularly useful in eliminating common noise between the two input voltages. For example, if )()()( 11 tntutv += , and )()()( 22 tntutv += where
)(1 tu and )(2 tu are signals of interest and )(tn is the common noise, then
7
)()()( 12 tututvout −= (when all resistors are equal) and the common noise is eliminated from the output signal. A differential amplifier can also be used to remove or add a DC offset to a signal. This is called “level shifting”. For example, again suppose all resistors are equal, Vtv 1)(1 −= , and )sin()(2 ttv ω= , then the output is the sine wave shifted up by 1 V DC. Example 7: Inverting summer Consider the circuit in Figure 9.
outV2V
2R
fR
2I
fI1V
1R
1I
3V3R
3I
−v
+v
Figure 9. Inverting summer We analyze the circuit, again using characteristics 1 and 2.
)(3
3
2
2
1
1
3
3
2
2
1
1
321
RV
RV
RVRV
RV
RV
RV
RV
IIII
fout
f
out
f
++−=
−=++
=++
If fRRRR === 321 , then outv is just the inverted sum of the inputs. A block diagram symbol for the inverting summer circuit is shown in Figure 10. A non-inverting summer could be created by following the inverting summer with an inverting op amp (Figure 3 with 21 RR = ).
8
2v
1v
3v
outv
Figure 10. Inverting summer Example 8: Another Summer Consider the circuit in Figure 11. Recall that all voltages are referenced to ground.
outV2V
2R
fR
2I
fI1V
1R
1I
4V4R
4I
3V3R
3I
−v
+v
Figure 11. Another Summer
f
ou t
f
RVv
RvV
RvV
III
−=
−+
−
=+−−−
2
2
1
1
21
)(),(
,
)()(
)(
43
3
43
4
43
443
33
43
4
4
4
3
3
43
RRR
BRR
RAwhere
BVAVv
VRR
RV
RRRv
RvV
RvVII
+=
+=
+=
++
+=
−−=
−
−=
+
+
++
9
Substituting, we get:
)1)((),1)((),(),(
,
)()()(
2143
3
2143
4
21
4321
43
2
432
1
431
+++
=+++
===
++−−=
−+=
+−+
+−
RR
RR
RRR
FRR
RR
RRRE
RR
DRR
Cwhere
FVEVDVCVVR
VBVAVR
BVAVVR
BVAVV
ffffff
out
f
out
A block diagram of this summer is shown in Figure 12.
2v
1v
3voutv
C
F
D
E
4v Figure 12. Block diagram of summer For example, if fRRRRR ==== 4321 , then the weights on the inputs are: C=D=1 and E=F=1.5. Example 9: Low-pass filter Consider the circuits in Figure 13.
outV
2R
1V1R
−v
+v
C
Figure 13. Low-pass filter
10
CRRRAwhere
jA
CRjR
R
ZZ
VV
CRjR
RCjZ
RZ
in
out2
1
2
2
1
2
1
2
2
2
1
22
11
,,11
11
==+
−=+
−=−=
+=⎥
⎦
⎤⎢⎣
⎡+=
=−
τωτω
ωω
For example, suppose FCRkR µ1,100,1 12 =Ω=Ω= , then msA 1,10 == τ and the Bode plot of the system is shown in Figure 14 below.
Figure 14. Bode plot of system in Figure 13. Bandwidth = 103 rad/s (160 Hz) Example 10: High-pass filter Consider the circuits in Figure 15.
outV
2R
1V1R
−v
+v
C
Figure 15. High-pass filter
11
CRCRwherejj
CRjCRj
RCj
RZZ
VV
RZ
RCj
Z
in
out2211
1
2
1
2
1
2
1
2
22
11
,,111
1
==+
−=+
−=+
−=−=
=
+=
ττωτ
ωτω
ω
ω
ω
For example, suppose ss 3
26
1 10,10 −− == ττ , then a Bode plot of the system is shown in Figure 16 below.
Figure 16. Bode plot of system in Figure 15. Cutoff frequency = 106 rad/s (160 kHz)
12
Example 11: Band-pass filter Consider the circuits in Figure 17.
outV
2R
1V1R
−v
+v
1C
2R
2C
Figure 17. Band-pass filter
,)1)(1()1)(1(
11
1
21
3
2211
12
1
2
22
21
222
11
1
ωτωτωτ
ωωωω
ω
ω
jjj
CRjCRjCRj
ZZ
VV
CRjR
RCjZ
RCj
Z
in
out
++−=
++−=−=
+=⎥
⎦
⎤⎢⎣
⎡+=
+=
−
123222111 ,, CRCRCRwhere === τττ For example, suppose 3
38
26
1 10,10,10 −−− === τττ , then the Bode plot for the system is as shown in Figure 18.
13
Figure 18. Bode plot of system in Figure 17. Bandwidth = 102 rad/s (16 Hz) Real Op Amps Real op amps have characteristics that are approximated by the ideal characteristics stated on page 1: Characteristics Characteristics that are useful in analyzing an op amp circuit are:
1. Current into +v and −v is small (~10-9 – 10-12 A). 2. Voltage at the two inputs is non-zero but approximately equal when op-amp is
configured with negative feedback (i.e., when the output is connected to the negative input terminal).
3. Input impedance looking into the input terminals is large: ΩΩ≈ 127 1010 toRin . The larger input impedances are found on op amps with FET input stages (like the TL081).
4. Output impedance is small: Ω≈ 60outR Figure 19 shows a model of an op amp. Internal circuitry of an op amp can be found in its data sheet. The open-loop gain (A) of an op amp is large (105 – 106). The open-loop gain is the gain when there is no connection between the output and the negative input terminal, as we have seen in the initial examples. The input is: −+ −= vvvin . The output is: )( −+ −== vvAAvv inout .
14
outv−v
+v
inR outR
inAvinv
Figure 19. Op amp model A/D Converters An A/D (analog to digital) converter converts an analog (continuous) signal to a digital (discrete) signal. Without feedback, an op amp amplifies the input signal until it saturates at either the positive or negative external supply voltage ( V15± ). A group of op amps can be used to convert an analog signal to a “digital” signal, where the output is either ON (+15V) or OFF (-15V). This is shown in Figure 20 for converting a 0-3 V signal to a digital signal with 1 V accuracy.
2.5 V
2R
2R
R
V3
V3
V1
1.5 V
0.5 V
Vin
Figure 20. A/D Converter
15
We define a “1” output as 15V and a “0” output as -15V. Table 1 below shows the output voltages for the respective input voltages. Note that the A/D converter can’t distinguish between a 2V input and 2.2V input. It only distinguishes between 1 V differences. A digital circuit could convert the 3-bit output to a binary representation. This is a 2-bit A/D converter. Vin (V) 0 1 2 3 V3 V2 V1 000 001 011 111 Binary 00 01 10 11 Table 1. A/D converter outputs Lab Hints
• Don’t forget to connect the power supplies ( V15± ) to each of your op amps before you try to use them.
• Test each op amp chip using a simple circuit (like a voltage follower) to confirm that the chip operates correctly.